ML20235R342

Insp Repts 50-324/88-200 & 50-325/88-200 on 880926-1007. Technical Inaccuracies Noted.Major Areas Inspected:Emergency Operating Procedures,Procedures & Plant Operations Re Human Factors & Primary Containment Venting Procedures

ML20235R342
Person / Time
Site:	Brunswick
Issue date:	01/25/1989
From:	Haughney C, Konklin J, Vandenburgh C Office of Nuclear Reactor Regulation
To:
Shared Package
ML20235R333	List: ML20235R331 ML20235R342
References
50-324-88-200, 50-325-88-200, NUDOCS 8903030263
Download: ML20235R342 (35)

See also: IR 05000324/1988200

Text

_ _ .

.

.

t

"

.

.

U.S. NUCLEAR REGULATORY COMMISSION

OFFICE OF NUCLEAR REACTOR REGULATION

Division of Reactor Inspection and Safeguards

Report Nos.:

50-325/88-200 and 50-324/88-200

Docket Nos.:

50-325 and 50-324

Licensee:

Carolina Power and Light Company

P.O. Box 1551

Raleigh, NC 27602

Inspection At:

Brunswick Steam Electric Plant, Units 1 and 2

Inspection Dates:

September 26 through October 7, 1988

l

~ b' ^-

Team Leader:

l -T5- @

-

C. A. VanDenburgh, Senior Operations

Date Signed

I

Engineer, NRR

.

l

,

Team Members:

G.T. Hopper, Region II

P.R. Farron, Nuclear Engineers and Consultants

D.H. Schultz, Comex Corporation

J.F. Hanek, EG&G Idaho, Incorporated

W.E. Gilmore, EG&G Idaho, Incorporated

Other NRC Personnel Attending Exit Meetings:

J. Konklin, Section Chief NRR;

C. Julian, Branch Chief, Region II; B. Buckley, Project Manager, NRR; and

W. Ruland, Senior Resident Inspector.

l

Reviewed By:

w

/

//ET//f'(

ames E. Konklin, Chief

Ddte Signed

'

Special Team Support

& Integration Section, NRR

Approved By:

(164/4t(

//2Md7

Unarles p. Haughney, Chief

Ddte Signed

Special inspection Branch, NRR

i

8903030263 890223

PDR

ADOCK 03000324

O

PDC

_ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _

-

-

.

'

,

~

.

Scope:

1

From September 26 through October 7,1988 an NRC inspection team conducted an

'

inspection of the Emergency Operating Procedures (E0Ps) for the Brunswick Steam

Electric Plant (BSEP), Units 1 and 2.

BSEP Units 1 and 2 are General Electric

i

BWR-4 plants with Mark I containments.

The objective of the inspection was to

determine if the E0Ps (1) were technically correct (2) could be physically

carried out in the plant, and (3) could be performed correctly by the

operators.

The inspection team compared Revision 4AF of the BWR Owner's Group (BWROG)

Emergency Procedure Guidelines (EPGs) to the Plant Specific Technical

Guidelines (PSTGs); compared the PSTGs to the E0Ps; reviewed the calculations

performed to develop the plant specific curves, values and setpoints utilized

in the E0Ps; performed a plant walkthrough of all the E0Ps and the Local

l

Emergency Procedures (LEPs) and Supplemental Emergency Procedures (SEPs)

'

referenced by the E0Ps; observed a simulation of four emergency scenarios using

the plant-specific simulator; performed a human factors review of the

procedures and plant operations; interviewed licensed and non-licensed

personnel who utilize the E0Ps; and reviewed the primary containment venting

procedures.

Results:

The inspection was based on a draft of the E0Ps which were in the final stages

of development and were expected to be implemented on December 15, 1988.

The

draft E0Ps incorporated Revision 4AF of the BWROG EPGs.

They corrected

i

l

deficiencies which had been identified during an Operational Safety Assessment

'

and a Probabilistic Risk Assessment based inspection [ Inspection Reports

50-325(324)/88-19 and 50-325(324)/88-11] performed by Region II to evaluate the

E0Ps presently in use.

The inspectors were impressed with the scope of the corrective actions taken in

response to the deficiencies identified during the previous inspections and

with the licensee's controls for the development of the E0Ps. All of the

previous deficiencies had been corrected, and the development process was well

documented and defined.

l

The BSEP E0Ps were developed as post-trip recovery procedures and integrated

the post-trip operator actions with the required actions of the EPGs and the

l

station blackout actions.

The E0Ps provided a high level of detail and

l

prioritized the operators' actions based on the significance of the event. As

a result, however, the E0Ps had a significant potential to delay the required

accident mitigation actions as post-trip recovery actions were accomplished.

The inspection team concluded, based on the simulator scenarios, that the

required EPG actions could not be accomplished in a timely manner without the

direct involvement of both the shift foreman and the shift technical adviscr to

read and perform the E0P action steps. The active participation of both these

individuals was not in accordance with the licensee's administrative

instructions, but was considered by the team to be an adequate method of E0P

accomplishment.

l

The inspection team determined that the draft E0Ps did not in every instance

represent an accurate incorporation of the BWROG EPGs and would not adequately

1

m___._______________.___.__..______________._

_ _ _ _ _ _ _ _ _

.

.

'

.

.

.

assure the successful accomplishment of all specified actions because several

procedures had a low probability of success and several calculational errors

were identified.

Several of the inspection teams' concerns affected the E0P3

which were presently implemented. The licensee was requested to take innediate

action to evaluate and correct these operational concerns.

J

l

l

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

,-.

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

_

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

- - - - . - _

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

.

.

t

.

-

.

TABLE OF CONTENTS

EMERGENCY OPEP.ATING PROCEDURE INSPECTION at

Brunswick Steam Electric Plant, Units 1 and 2

(Inspection Reports 50-325/88-200and50-324/88-200)

Page

1.0 INSPECTION 0BJECTIVE.........................................

1

2.0

BACKGR0VND.......................................

4

1

..........

3.0 DETAILED INSPECTION

FINDINGS.................................

3

3.1 Emergency Operation Procedure (E0P) Program Evaluation..

3

3.1.1

E0P Development..................................

3

3.1.2 Licensee Verification and Validation of E0Ps.....

4

i

3.1.3

E0P Operator Training............................

5

1

3.1.4 Maintenance of E0Ps..............................

6

3.1.5 Quality Assurance Involvement in PSTG

l

1

Maintenance......................................

6

3.1.6 Licensee Response to IE Information Notice 86-64.

7

3.2 E0P P rocedu re Ve ri fi ca t i on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

!

3.2.1

EPG/PSTG Comparison..............................

7

I

3.2.2 PSTG/EOP Comparison..............................

12

3.2.3 Calculation Review...............................

13

3.2.4 Adequacy of Writer's

Guide.......................

15

3.2.5 Writer's Guide Implementation....................

16

3.3 E0P Validation Using Plant Walkthroughs.................

17

j

3.3.1 Technical Adequacy of Procedures.................

18

3.3.2 Availability of Special Tools and Equipment......

19

3.3.3 Station Material

Condition.......................

21

3.3.4 Reactor Building Accessibility...................

21

l

3.4 C0P Validation Using Plant Simulator....................

22

3.4.1

Scena rio Des c ri pti on. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

22

,

3.4.2 Limitations of the Plant-Specific Simulator......

23

'

3.4.3 Observations and Conclusions.....................

24

'

3.5 Operator Interviews.....................................

25

3.5.1 Observations and Conclusions.....................

26

3.6 Prima ry Contai nment Venting Provi sions. . . . . . . . . . . . . . . . . .

27

4.0 MANAGEMENT EXIT MEETING......................................

27

Appendix A - PERSONNEL C0NTACTED..................................

A-1

Appendix B - DOCUMENTS REVIEWED...................................

B-1

_ _ .

--

.

- .

. .

!

-

.

'

.

.

.

1.0 INSPECTION OBJECTIVE

l

'

A special team inspection reviewed the licensee's Emergency Operating

Procedures (E0Ps), operator training and plant systems in accordance with NRC

Temporary Instruction (TI) 2515/92 to accomplish the following objectives:

1)

Determine whether the E0Ps conformed to the BWP Owner's Group (BWROG)

Emergency Procedure Guidelines EPGs) and were technically correct for the

BrunswickSteamElectricPlantg{BSEP), Units 1and2.

,

2)

Assess whether the E0Ps could be physically carried out in the plant using

existing equipment, controls, and instrumentation, under the expected

environmental conditions.

3)

Evaluate whether the plant staff could correctly perform the E0P actions

in the time available.

2.0 BACKGROUND

Following the Three Mile Island (TMI) accident, the Office of Nuclear Reactor

Regulation (NRR) developed the "TMI Action Plan," (NUREG-0660 and NUREG-0737).

Item I.C.1 of this plan required licensees of operating plants to reanalyze

transients and accidents and to upgrade E0Ps.

In addition, Item I.C.9 of the

plan required the NRC staff to develop a long-term plan that integrated and

expanded efforts for the writing, reviewing, and monitoring of plant

procedures.

NUREG-0899, " Guidelines for the Preparation of Emergency Operating

and describes the use of a Procedures Generition Package (PGP) pgrading E0Ps,

Procedures," represents the NRC staff's long-term program for u

to prepare E0Ps.

l

The licensees formed four vendor owners groups corresponding to the four major

reactor vendor types in the United States: Westinghouse, General Electric,

Babcock & Wilcox, and Combustion Engineering. Working with the vendor

companies and the NRC, the owner's groups developed generic procedures that set

forth the desired accident mitigation strategy.

For General Electric plants,

the generic guidelines are referred to as the BWR0G EPGs. These guidelines

were to be used by the licensees in developing their PGPs.

Generic Letter 82-33, " Supplement I to NUREG-0737 - Requirements for Emergency

Response Capability," required each licensee to submit to the NRC a PGP which

included, (1) Plant Specific Technical Guidelines (PSTGs) with justification

,

for safety significant differences from the BWROG EPGs, (2) a Plant Specific

Writer's Guideline (PSWG), (3) a description of the program to be used for the

verification and validation of E0Ps, and (4) a description of the training

program for the upgraded E0Ps. The generic letter required the development of

plant-specific E0Ps which would provide the operators with directions to

mitigate the consequences of a broad range of initiating events and subsequent

multiple failures or operator errors.

In addition, the upgraded E0Ps were

required to be symptom-based procedures which would not require the operators

j

to diagnose specific events.

I

Although various circumstances caused long delays in achieving NRC approval of

J

many of the PGPs, the licensees have all implemented their upgraded E0Ps.

To

determine the success of this implementation, a series of NRC inspections was

performed to examine the final product of the program - the E0Ps.

!

i

1

!

.

.

-

l

l

Representative samples of each of the four vendor types were selected for

j

review by four inspection teams from Regions I, II, III and IV.

An additional 13 inspections were identified at facilities with General

Electric Mark I type containments. These inspections were conducted by the

Office of Nuclear Reactor Regulation and included a detailed review of the

primary containment venting provisions of the E0Ps.

This inspection is the

final inspection in this series.

3.0 DETAILED INSPECTION FINDINGS

j

I

3.1 Emergency Operating Procedure (EOP) Program Evaluation

l

3.1.1

E0P Development

A Confirmatory Order dated February 22, 1984, identified that the licensee had

submitted a PGP on August 17, 1983, and implemented upgraded E0Ps.

The PSTG

submitted and the E0Ps currently implemented at the facility were based upon

l

Revision 2 of the BWROG EPGs.

l

l

The inspection team reviewed a draft version of the E0Ps which were based upon

Revision 4AF of the BWROG EPGs.

This revision incorporated a revised accident

mitigation strategy and calculational methods which were approved by the NRC in

a generic safety evaluation report (SER) issued on September 12, 1988.

The

inspection team reviewed the draft E0Ps because the licensee was in the final

stages of implementing this revision and had scheduled full implementation by

December 15, 1988.

Although the inspection was based upon the draft E0Ps, the

l

inspection team verified whether identified deficiencies affected the approved

'

ECPs.

Two operational concerns were identified and are discussed in Sections

3.2.1.1 and 3.2.1.2 of this report.

Both the currently implemented E0Ps and '.he draft E0Ps had been developed in

flowchart format with the post-trip recovery actions and the station blackout

actions integrated with the steps of the BWROG EPG accident mitigation

strategy. The post-trip recovery actions are event-based actions which are

normally provided in separate procedures and are not appropriate for the

symptom-based E0Ps.

The BWROG EPGs and the SER indicated that additional

auxiliary event-specific procedures intended for use in conjunction with the

symptomatic procedures must not contradict or subvert the symptomatic operator

actions specified in the BWROG EPGs.

The inspection team was concerned that the inclusion of the event-based actions

into the E0Ps delayed the accomplishment of the actions directed by the BWROG

EPGs, had the potential to result in incorrect event diagnosis, and affected

l

the ability of the operators to implement the E0Ps and thereby respond to the

emergency in a timely manner. As discussed in Section 3.4.3.1, the simulator

scenarios demonstrated that the shift foreman could not implement the E0Ps, as

required by the licensee's administrative procedures, without the assistance of

the shif t technical advisor to directly monitor and control the specified E0P

actions involving primary containment and radiological release control.

Although operation in this manner was not in accordance with the administrative

procedures, the inspection team concluded that the operating crew could

implement the specified E0P actions to shutdown the reactor and return the

plant to a safe, stable condition. However, the inspection team identified

2

_ _ _ -

_-

_ - _ _ _ _ - _

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

.

.

.

-

.

several examples as a result of this method of implementation, in which

specified E0P actions were not accomplished or were misinterpreted because the

shift foreman and the STA were involved in separate areas of the E0Ps and did

not have the opportunity to consult and review each others actions.

The inspection team identified the following additional examples in which the

E0Ps included event-based actions not related to post-trip recovery actions.

1)

Path-2, steps 35, 60 and 170, precluded the use of the feedwater or

condensate system for reactor pressure vessel (RPV) injection if the

feedwater conductivity was greater than 0.3 mmhos. This was an

event-based action for condenser tube leakaoe which potentially delayed or

prevented recovery from a low RPV water level condition.

2)

Primary Containment Control Procedure, steps PC/P-19 through 22, were

event-based actions for recirculation pump seal failure which were not

related to primary containment control.

3)

In the Radiological Release Control Procedure, all the steps in the five

flowpaths below step RR-6 were event-based actions for identification and

mitigation of primary leakage. Although these steps were necessary in the

event of a primary leak, they wv.e not specified in the BWROG EPG for the

response to a radiological release.

Further licensee action is necessary to ensure that the E0Ps do not contain

event-based actions and to implement the E0Ps in a manner consistent with the

administrative procedures.

,

In addition to including event-based actions in the E0Ps, the licensee

developed the E0Ps with a high level of detail and complexity.

The inspection

team was concerned that the additional detail and unnecessary complexity

represented by the following examples, had the potential to delay the

i

operator's response to an actual emergency.

Further licensee action'is

I

necessary to reduce the level of complexity of the E0Ps.

l

l

1)

Primary Containment Control Procedure, step C.9.e (1), and section 1,

!

included reference to the head spray system, which was no longer

'

applicable because of a plant modification.

The reference should be

'

deleted.

2)

Primary Containment Control Procedure, step PC/M-4, indicated that the

hydrogen monitor readings must be compensated for primary containment

conditions in accordance with operational procedure OP-24.

In practice,

as evidenced by the simulator scenarios, the operators did not consult

OP-24 to determine correction values and there was inadequate time to

perform these calculations.

3)

Path-2 and Path-3 provided multiple steps for initiating suppression pool

cooling. As demonstrated in the simulator exercises, the operators placed

suppression pool cooling into operation without working through each of

these steps.

By contrast, the SP/T path of primary containment control

for operating suppression pool cooling provided direction to the operator

using only a single step.

3

'

.

'

,

.

.

4)

Path-2 was entered following a reactor scram from a condition where the

reactor mode switch was not in RUN. The power level at this condition is

anticipated to be less than approximately eight percent power.

Steps 12,

13 and 17 of this procedure represented actions for tripping the main

turbine, ensuring that turbine auxiliaries started, and tripping the

heater drain pumps. These actions were not appropriate for this power

level and diverted the operators attention from more important activities.

5)

Path-3, step 26, required the operator to set the reactor re irculation

pump speed controllers to minimum; however, this step was not required for

the pump logi: involved.

6)

Radiological Release Control Procedure, steps RR/PB-9 through 11,

identified core cooling systems which -ay be the source of a primary leak;

,

however, these systems were not located in the turbine building and were

therefore not applicable.

7)

Path-3, steps 61 and 85, and Path-2, steps 61 and 85, provided redundant

action steps for RPV pressure control.

P)

Throughout the E0Ps, there were several examples in which multiple action

steps were used to accomplish a single action.

For example, in each leg

of the Primary Containment Control Procedure, the generic monitor and

control steps of the PSTG were restated in one step and the specific

direction on how to accomplish the referenced action was provided in the

subsequent step.

In addition, if a scram was required during the

performance of the Primary Containment Control Procedure, two action steps

were required. The first stated that a reactor scram was required and the

second executed the scram.

,

l

9)

Path-2, steps 80, 90, 91, 99 and 100, and Path-3, steps 79 through 84,

j

provided detailed steps for the operation of the suppression pool cooling

system which were more appropriate for operational procedures.

In

,

addition, these actions should have been covered in a single action step

as accomplished by procedure SP/T-3.

.

3.1.2 Licensee Verification and Validation of E0Ps

i

NUREG-0899, section 3.3.5, indicated that after E0Ps are written they must

i

undergo a process of verification and validation. This process was used to

l

establish the accuracy of information and instructions, to determine that the

'

procedures could be carried out accurately and efficiently, and to demonstrate

that the procedures were adequate to mitigate transients and accidents.

Both

technical and human engineering adequacy were required to be addressed in the

review process.

i

Administrative Instruction AI-95, " Verification and Validation Program for EPG,

Revision 4, based Emergency Operating Procedures," defined the program for

verification and validation of the E0Ps at BSEP.

BSEP administrative

procedures (Volume 1, Book 1, sectic' 5.7.4.1, paragraph A.III.c) required that

all E0P changes receive the review and approval of the E0P Review Committee.

i

The membership of the committee included operations, quality assurance,

technical support, off-site nuclear safet.;, and the licensed training

!

departments.

4

!

!

'

.

.

4

j

lhe verification and validation required for implementation of Revision 4AF of

the BWROG EPGs was identified in an untitled supplement to AI-95. The licensee

indicated that this methodology was reviewed and approved by the E0P Review

Committee on December 14, 1987.

The team was concerned that the specific

methodology for the verification and validation of the substantial changes

represented by the incorporation of Revision 4AF of the BWROG EPGs had not been

specifically identified and approved by the E0P Review Committee.

Further

licensee action is necessary to ensure that the verification and validation

program approved by the E0P Review Comittee is successfully completed prior to

the implementation of the draft E0Ps.

The inspection team's review of the verification and validation program

determined that a mechanism existed for ensuring that all portions of the E0Ps

could be validated using either the plant-specific simulator, plant

l

walkthroughs, or desk top reviews. The preferred validation method was to

perform the E0Ps on the plant simulator.

In instances in which the E0P steps

exceeded the capability of the simulator, a combination plant walkthrough and

desk top review was employed.

In order to ensure that the full complement of

E0Ps were validated, a list of functional objectives to be accomplished by

performing the E0P was developed. The licensee defined the functional

objective of each E0P step and developed exercises to satisfy each functional

objective. The exercises were performed on the plant simulator or by some

combination of simulator exercises and plant walkthroughs.

Problems idantified

during the demonstration of the functional objectives were resolved by the E0P

revision process as described in Al-95.

l

The inspection team was concerned that the identified verification and

I

validation methodology did not indicate that appropriate consideration had been

l

given to the necessity of performing all steps of the E0Ps on the plant

simulator or in table top exercises rather than evaluating them through plant

walkthroughs. Based on the inability of the simulator to accurately model RPV

level and decay heat (previously identified by the licensee and discussed in

Section 3.4.2), it is extremely important to ensure that all steps of the E0Ps

would be effective in fulfilling the actions intended by the BWROG EPGs.

Further licensee action is necessary to ensure that all E0P steps are validated

on a simulator or by an equally acceptable methodology.

3.1.3

E0P Operator Training

A review of the licensee's training program was conducted to determine the

adequacy of operator training prior to implementation of Revision 4 of the

.

EWP0G EPGs.

The inspection team compared the requirements of NUREG-0899, the

l

Procedures Generation Package (PGP), and the operator training program

l

developed for Revision 4.

The PGP contained a detailed description of the initial operator training which

was conducted prior to the initial implementation of the upgraded E0Ps. The

extent of the initial operator training met the requirements of NUREG-0899,

section 3.4

Although the PGP did not specify the operator training require-

ments for revisions to the E0Ps, the licensee required all revisions to be

performed in accordance with Administrative Instruction Al-95, " Verification

and Validation Program for EPG, Revision 4, based Emergency Operating

Procedures." This procedure required that an E0P Review Comittee review

proposed E0P revisions and determine the implementation requirements.

5

___- _ _ _ _ -

_____-

'

.

'

.

.

The inspection team was concerned that these instructions contained no guidance

concerning the scope of operator training required prior to implementation of

revisions to the E0Ps.

Nevertheless, the licensee developed and incorporated a

satisfactory training program into the E0P verification and validation process.

This program was outlined in an untitled supplement to AI-95. The licensee

recognized the need for ongoing operator training on the E0Ps and had

accomplished this goal with periodic licensed operator retraining and the

,

Operator Real Time Training Program described in Operating Instruction 01-33.

This latter program served to complement the annual operator retraining by

accomplishing immediate training needs on a continuing basis.

Interviews with training supervisors indicated that two of the three phases of

training had been completed prior to the time of the inspection. Phase IA was

completed in April 1988 and consisted of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of classroom instruction on

i

the content and use of the new E0Ps.

Phase IB was completed in June 1988 and

l

consisted of 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> of classroom instruction cambined with 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of

simulator training designed to exercise the major branching points within the

E0P flowcharts.

Phase 11 training was conducted in September 1988 and involved

l

four hours of classroom instruction followed by four hours of simulator

exercises.

This brief session served to update operators on changes made to

the E0Ps since completion of Phase IB training.

The final phase of training

(Phase III) was accomplished in December 1988 and served as the final operator

training update prior to implementation of the revised E0Ps.

3.1.4 Maintenance of E0Ps

During the review of the PSTGs and the E0Ps, the team determined that the PSTGs

j

and BWROG EPG Appendix C calculations were being maintained up-to-date as a

basis document ar.d were properly controlled as a plant record by the document

control center.

The E0P calculations based on Revision 2 of the BWROG EPGs

were reviewed and documented in a study entitled ENSA 84-038, "EOP Numerical

Limits and Graphs," and the PSTG was contained in Operating Instruction 01-37,

" Preparation and Review of the Plant Specific Technical Guideline for EPG

Revision 2."

The E0P calculations based on Revision 4 of the BWROG EPGs were

under review by the Nuclear Engineering Department and were scheduled to be

published and the PSTG documented in a similar manner.

3.1.5 Quality Assurance Involvement in PSTG Maintenance

NUREG-0899, section 4.4, indicated that as a primary basis of the E0Ps, the

PS1Gs should be subject to examination under the plants' overall quality

assurance (QA) program.

Because the licensee was responsible for ensuring that

the PSTGs were accurate and up-to-date, the review and control of the PSTGs

shculd be included in the established QA program.

The licensee indicated that QA surveillance 86-067 was performed in December

1986 as a result of E0P development deficiencies identified by the NRC in IE

Information Notice (IEN) 86-64.

In addition, QA Audit QAA/0021-88-05, was

performed in August 1988 on Revision 4 of the E0Ps and identified one follow-up

item concerning justification 'of BWROG EPG deviations.

Future audit schedules

included a site QA surveillance, similar in scope to surveillance 86-067,

scheduled for the first quarter of 1989 and annually thereafter.

6

_

___ _

-

-

_ - _ _ _ _ _ _ _ _ _ _ _ _ _ _

'

.

'

.

.

.

3.1.6 Licensee Response to IE Information Notice 86-64

IEN 86-64 was issued on August 14, 1986, followed by IEN 86-64, Supplement 1,

issued on April 20, 1987.

IEN 86-64 alerted the licensee to problems found in

review and audits of Procedure Generation Packages (PGPs) and E0Ps. The IEN

indicated that many utilities had not appropriately developed or implemented

upgraded E0Ps.

In addition, the IEN identified deficiencies in the development

and implementation of each of the four major aspects of the upgrade program.

These deficiencies included undocumented deviations from and inappropriate

adaptation of BWROG EPGs, failure to adhere to the PSWG and the verification

and validation programs, and deficient training programs.

Supplement 1 to IEN

86-64 alerted the licensee's to significant problems that were continuing with

plant E0Ps. Deficiencies were identified in all the major aspects of the E0P

upgrade program.

The licensee's were requested to review the information for

applicability to their facility and consider actions to correct or preclude

similar problems from occurring.

The licensee's evaluation process for IENs was performed in accordance with

Corporate huclear Safety Instruction CNSI-I and On-Site Nuclear Safety

Instruction ONSI-1.

The IENs were reviewed by the nuclear safety coordinator

,

and assigned to responsible engineers for evaluation.

IEN 86-64 and Supplement

1 were evaluated by the coordinator and closed because a OA surveillance,

'

discussed in Section 3.1.5, had already been initiated and had identified

similar deficiencies. The inspection team concluded that the licensee's

actions in response to IEN 86-64 were satisfactory.

.

3.2 E0P Procedure Verification

This portion of the inspection was performed to determine whether the E0Ps had

been prepared in accordance with the BWROG EPGs, the PSTGs, and the PGP.

l

The inspection ccmpared Revision 4AF of the BWR0G EPGs to the PSTGs, and the

PSTGs to the E0Ps.

All differences were evaluated to ensure that safety

significant deviations were identified and that a documented basis existed for

all deviations. A review of selected calculations was performed to ensure that

plant-specific values utilized in the E0Ps were correct and had been performed

in accordance with a documented engineering analysis. Appendix B of this

report lists the procedures reviewed.

3.2.I

FPG/PSTG Comparison

Nine differences were identified between the BWROG EPGs and the PSTGs as

detailed below.

Based on these discrepancies, the inspection team concluded

that the draft PSTGs did not accurately incorporate the guidance of Revision

4AF of the BWROG EPGs. The inspection team identified technical concerns

relating to the measurement of RPV water level, which adversely affected the

I

operator's ability to perform the level / power control procedure, and technical

I

concerns relating to the measurement of primary containment drywell temperature,

which potentially raasked a valid entry condition.

These concerns affected both

the E0Ps which were currently implemented at the facility (Revision 2) and the

draft E0Ps.

In addition, numerous discrepancies were identified in the draft

i

'

E0Ps in which the entry conditions of the PWROG EPGs were changed without

sufficient technical justification.

Further licensee action is necessary to

l

evaluate and correct the E0Ps presently in use and to ensure that the draft

!

E0Ps accurately incorporate the technical guidance of the BWROG EPGs.

7

. _ _ - - _ _ _ _

___

_

,

'

.

S

-

.

1)

BWROG EPG Contingency No. 7 provided a methodology to control reactor

power following an anticipated transient without scram (ATWS).

This

methodology involved lowering the RPV water level to the top of active

fuel (TAF) or to the minimum steam cooling water level (MSCWL).

The

licensee implemented these actions in the Level / Power Control Procedure.

BWROG EPG Caution ho. 1 provided operator precautions related to the

nieasurement of RPV water level and the accuracy of various water level

instruments. The licensee implemented these precautions in Caution No. 1

of the User's Guide. The inspection team reviewed the methodology and

precautions for ATWS power level control implemented by the licensee and

identified several undocumented and unjustified deviations which adversely

affected the ability of the operators to control reactor power.

These

deviations involved (1) the equivalency between the instrument zero

indication and the TAF, (2) the restrictions on the use of the wide range

level instruments, and (3) the calibration of the fuel zone level

instrument.

These conditions, detailed in the following paragraphs,

affected the E0Ps which were presently in use at the facility.

a)

Instrument Zero - Based on the location of the instrument taps, the

wide range level instruments (N0-26A and NO-26B) indicate 0 inches

when the actual RPV water level is +8.44 inches above the top of

active fuel (TAF).

In an attempt to simplify the E0Ps, the licensee

,

used this instrument zero indication as the TAF; however, the

licensee did not document this deviation from the BWROG EPGs and did

not evaluate the difference to ensure that there were no adverse

effects on the Level / Power Control Procedure. This deviation was

significant because water levels higher than the TAF result in higher

reactor power levels during an ATWS condition.

Higher power levels

increase the amount of energy deposited in the primary containment

and reduce the time until primary containment venting is t equired.

In addition, a larger primary containment vent path may bt. required

to remove this excess energy. The inspection team also noted that

this condition affected the E0Ps which were presently in use at the

facility.

Further licensee action is necessary to ensure that the

higher power levels resulting from this deviation are technically

acceptable and appropriately documented for the approved and draft

E0Ps.

b)

Wide Range Level Instrument Restrictions - Caution No. 1 in the

User's Guide restricted the use of the wide range level instruments

(N0-26A and NO-268) as a function of level. The caution required

that the instruments not be used when the indicated water level was

below +10 inches (i.e., +18.44 inches actual) on both Units 1 and 2.

In addition, the caution precluded use of the Unit 2 instruments when

the water level was below +40 inches (i.e., +48.44 inches actual)

when conditions indicative of a high energy line break (HELB) were

present.

These restrictions were based upon the location of the

reference le.gs of the wide range instruments and the lack of

temperature compensation methods in the E0Ps.

The +40 inch

precaution was not applicable on Unit 1 because the reference legs

were in a different location.

The licensee had not developed a method to compensate the level

instruments when indication was below +10 inches and did not have a

8

'

.

'

.

+

.

method to compensate the instruments during a HELB because

temperature instruments were not installed in the secondary

containment. The level instrument restrictions adversely affected

the performance of the Level / Power Control Procedure because the

operators did not have an accurate level instrument with which to

control the RPV level below an actual level of +18.44 inches or

+48.44 inches. This potentially degraded the control of reactor

power during an ATWS condition and, as such, was an undocumented and

unjustified deviation from the BWROG EPGs. The inspection team also

noted that these conditions affected the E0Ps which were presently in

use at the facility.

Further licensee action is necessary to

evaluate this deviation from the BWR0G EPGs and to provide an

accurate method to control reactor power by means of water level

during the ATWS condition for the approved and draft E0Ps.

c)

Fuel Zone Level Instruments Calibration - The fuel zone level

instruments (N0-36 and NO-37) were calibrated under cold conditions

of 212 degrees F in the reactor building and the drywell, and 0 psig

in the RPV. Under cold conditions, these instruments normally

indicate accurately from -150 to +150 inches. However, this cold

calibration resulted in a wide variance in actual versus indicated

level for the drywell temperatures, RPV pressure, and reactor

building temperature anticipated during an ATWS.

Because no

compensation method was available to the operators, the fuel zone

instruments would be grossly inaccurate under the conditions in which

they will be required to be used.

The following level deviations

would result if the fuel zone indicators were used at 1100 psig

during ATWS conditions in accordance with the Level / Power Control

Procedure. With actual RPV water level at the actual TAF (i.e.,

-8.44" indicated on the wide range instruments), RPV pressure,

reactor building temperature at 200 degrees F, and drywell

temperature in the area of the reference legs as indicated below, the

fuel zone instruments would indicate the following levels.

Drywell Temperature

Indicated Level

l

(degrees F)

(inches)

1

180

a,09

l

200

-61.45

250

57.06

300

-51.94

400

-39.60

500

-23.24

Since both units precluded the use of the wide range level

instruments below an indication of +10 inches, the fuel zone

indicators would be indicating approximately -42 inches below TAF at

the time when they became the only level indicators available.

The inspection team also noted that the E0Ps and associated cautions

did not preclude the use of the fuel zone instruments in preference

to the wide range instruments for water level control.

If the wide

range instruments were not available, the operators were required to

use the fuel zone instruments to control RPV water level. Under

I

9

_ - _

_ _ _ -

1

j

a

'

.

-

.

these conditions, and in the absence of compensation techniques, the

operators would control RPV water level to an indicativa of TAF

(i.e., an indication of 0 inches on the fuel zone instruments),

which would correspond to an actual RPV water level of approximately

+60 inches. Control of reactor power during an ATWS would not be

effective at these elevated levels.

The licensee's failure to provide a method of compensating the fuel

zone instruments for use in conditions other than their calibration

condition effectively prevented their use and had the potential to

adversely affect the performance of the Level / Power Control

Procedure.

This was a significant deviation from the BWROG EPGs

which was not documented or justified.

The inspection team also

noted that this condition affected the E0Ps which were presently in

use at the facility.

Further licensee action is necessary to provide

an effective method of controlling water level under the conditions

when use of the Level / Power Control Procedure is anticipated.

2)

The BWROG EPG drywell temperature entry condition was established at the

drywell technical. specification (TS) limiting condition for operation

(LCO) or the maximum normal operating temperature, whichever was higher.

The PSTG entry condition was set at the primary containment volumetric

average temperature LC0 of 135 degrees. The BSEP TS did not contain a LC0

for drywell temperature.

The PSTG justified this deviation based on the

assuinption that the values for drywell temperature LC0 and primary

containment temperature LC0 were equivalent even though the primary

containment volumetric average included the suppression pool air space in

addition to the drywell airspace.

Because the suppression pool air space

contributed 43 percent to the volumetric average of the primary contain-

ment, the potential existed for the relatively cool suppression pool air

temperature to mask a high temperature in the drywell.

In addition, the

PSTG justification indicated that the normal maximum operating temperature

was lower than 135 degrees; however, there were times throughout the year

when the maximum operating temperature exceeded 135 degrees.

The inssec-

tion team determined that this deficiency also affected the E0Ps whici

were presently in use at the facility.

Further licensee action is

necessary to correct this operational concern.

3)

The BWROG EPG entry conditions for the RPV control guideline were: (1) RPV

water level at the low level scram setpoint, (2) RPV pressure above the

scram setpoint, (3) drywell pressure above the scram setpoint, and (4)

reactor power above the average power range monitor (APRM) down scale trip

for any scram.

The entry conditions in the PSTG deviated from the BWROG

EPCs in that the PSTG entry condition was any plant condition requiring or

causing a scram.

The PSTG justification stated that this conservative

approach permitted execution of any of five scram recovery paths which

would lead the operator to the End Path Procedure where the entry

conditions of the BWR0G EPG would be assessed.

The inspection team was concerned that this methodology delayed essential

operator actions. The potential existed for plant parameters indicative

of an emergency (i.e., the BWROG EPG entry conditions), to remain

unmonitored and therefore uncontrolled pending completion of the post-trip

actions. These post-trip actions were event-based and are normally

10

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_

_

_ _ _ _ .

,

'

o

.

4

-

.

controlled as immediate, memorized actions' of the control room operators.

As discussed in Section 3.1.1, the inspection team was concerned that the

inclusion of these post-trip recovery actions into the E0Ps delayed the

accomplishment of the directed actions of the BWROG EPGs, had the

potential to result in incorrect event diagnosis, and affected the ability

of the operators to implement the E0Ps and thereby respond.to the

i

emergency in a timely manner. The licensee's method of satisfying the

1

BWROG EPG entry conditions and including event-based actions in the E0Ps

was a significant deviation from the BWROG EPGs and had the potential to

adversely affect the satisfactory performance of the E0Ps.

4)

The BWROG EPG entry condition for drywell pressure was the high drywell

pressure scram setpoint. The PSTG entry condition for drywell pressure

was established at the maximum pressure allowed by the plant technical

specifications of 2.0 psig, while the actual scram setpoint was 1.83 psig

+/- 0.076 psig. This deviation was not justified and was potentially

significant because a scram could occur at a high drywell pressure before

the Primary Containment Control Procedure entry conditions were satisfied.

5)

The BWROG EPG entry condition for the Radiological Release Control-

Procedure was limited to an ALERT condition from a radioactivity release

off-site. The PSTG entry conditions were more conservative than the BWROG

abnormal operating procedures (A0Fs) y conditions and actions for several

EPG entry conditions because the entr

were incorporated into the PSTG. As

discussed in Section 3.1.1, these additional actions diverted the

attention of the shift foreman during the simulator demonstration and

increased the complexity of the E0Ps.

6)

BWROG EPG, step C6-3, vented the RPV to permit flooding of primary

containment with a flow path through the RPV.

The specified vent paths

prevented pressurizing the primary containment during the Primary

Containment Flooding Procedure.

PSTG, step C6-2, improperly listed the

reactor head vent valves which vented to the floor of the primary

containment drywell. The vent lines did not accomplish the intent of the

BWROG EPGs because they were only 1/4 inch in diameter and were directed

inside primary containment. The same problem was noted at step C.7.a of

the Primary Containment Flooding Procedure.

7)

The BWR0G EPG entry condition for primary containment hydrogen

concentration was the high alarm setpoint for hydrogen concentration

(i.e., 2 percent). The PSTG entry condition was set at the minimum

detectable hydrogen concentration of 1 percent.

This value was

conservative with respect to the alarm, but relied on the operators to

monitor the concentration in order to identify the entry condition.

During an emergency this entry condition could be missed and could

potentially delay the operator actions required to mitigate the emergency.

8)

BWROG EPG, step C2-1.4, performed an emergency depressurization of the RPV

with other steam-driven equipment if the proper number of safety relief

valves (SRVs) could not be opened. The PSTG did not reference equipment,

such as the reactor feed pump turbines and steam jet air ejectors which

were also available at BSEP as additional steam loads capable of reducing

RPV pressure.

11

-

'

.

'

.

.

9)

BWROG EPG, step RC-1, required a manual scran of the reactor if a reactor

scram has not been already initiated.

The corresponding PSTG step

deviated from the BWROG EPGs by deleting this conditional action.

In the

justification for the deviation, the licensee indicated that the

conditional statement was deleted because the flowcharts were entered for

the initial scram and were not re-entered for any subsequent scrams.

The

inspection team was concerned that re-entry into the flowcharts would be

required if plant conditions changed and a new entry condition occurred.

!

Under these conditions, re-insertion of a scram signal was undesirable and

could adversely affect ongoing recovery actions such as alternate rod

insertion techniques.

l

3.2.2 PSTG/EOP Comparison

Four differences were identified in which the PSTGs steps were not accurately

incorporated into the E0Ps and were therefore unjustified deviations from the

BWROG EPGs.

Further licensee action is necessary to accurately incorporate

these PSTG steps.

1)

Paths 1, 2, 3, 4, and 5 included conditional action steps which precluded

the use of the feedwater system in the event of high condensate

conductivity.

These actions were not included in either the BWROG EPGs or

the PSTGs. The effect of these steps was to prevent the use of an

available high pressure injection system during a low RPV water level

emergency.

In addition, Path 5 failed to consider the use of the

feedwater system as a high pressure injection source until after the high

pressurecoreinjection(HPCI)andreactorcoreisolationcooling(RCIC)

systems were attempted. The SWROG EPGs assumed that the feedwater system

would be the first and primary method of level restoration, regardless of

the condensate conductivity, until after the RPV water level emergency was

controlled. The prerequisites for use of the feedwater system and the

failure to attempt its use are considered to be significant deviations

from the BWROG EPGs.

2)

PSTG, step RC/P-2, contained a conditional action step which placed the

control switch for each SRV in the CLOSE or AUTO position if the

continuous SRV pneumatic supply became unavailable. The intent was to

reserve operating air for subsequent necessary cycles of the SRVs.

PSTG,

step RC/P-3, required emergency RPV depressurization with sustained

opening of the SRVs if one or more SRVs were being used to depressurize

the RPV and the continuous SRV pneumatic supply became unavailable. The

intent was to continue the cooldown by leaving the appropriate valves open

continuously to maintain the proper cooldown rate.

However, the E0Ps in

End Path Procedure, step 76, required using the SRVs for RPV pressure

control only when a continuous pneumatic supply was available to the SRVs.

This was a deviation from the BWROG EPGs, in that sustained opening of the

SRVs was not attempted before operating pressure of the emergency

depressurization system was no longer available.

3)

PSTG, step DW/T-1, directed the operators to operate all available drywell

cooling, defeating isolation interlocks if necessary.

However, the

Primary Containment Control Procedure, step DW/T6, prohibited operation of

the drywell coolers if drywell pressure was above 2.0 psig. The licensee

indicated that the operation of the drywell coolers was prohibited at 2.0

12

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - -

.

'

.

'

.

psig because the fans had previously tripped on thennal overloads at this

pressure.

This rationalization did not justify the restriction on the

primary method available to mitigate the high drywell temperature

condition.

Further licensee action is necessary to investigate and

correct the drywell cooler fan problems in order to fully utilize the

drywell coolers for primary containment temperature control.

4)

PSTG, step RC/P-1, directed the operators to manually open the SRVs if any

SRVs were cycling, until reactor pressure dropped to 950 psig, the

pressure at which all turbine bypass valves would remain fully open.

However, E0P Path-1, step 12, required that the operators open SRVs to

stabilize reactor pressure while maintaining maximum possible steam flow

to the main condenser, and did not specify a pressure setpoint. The

inspection team was concerned that the E0P omitted the parameter to which

the RPV pressure should be lowered without justification.

3.2.3 Calculation Review

l

The inspection team reviewed the calculations for figures and setpoints used in

the E0Ps to determine if the values were correctly calculated based on the

plant specific differences and the guidance of the BWROG EPGs. At the time of

the inspection, the licensee's Nuclear Engineering Department (NED) was

completing an independent verification of all calculations used to support the

draft E0Ps in accordance with Special Procedure SP-87-079, Revision 001,

.

" Independent Review of BSEP E0P Numerical Limits and Graphs." Although several

l

calculations remained to be verified by the NED, the calculations reviewed by

'

the inspection team had previously been completed by the NED. As evidenced by

1

the errors in the calculation of the hot shutdown boron weight discussed below,

the verification of the draft E0P calculations was not completely effective.

Further licensee action is necessary to ensure the accuracy of the calculations

and associated assumptions. The following deficiencies were noted.

1)

Worksheet WS-09 determined the maximum primary containment water level

limit that would not cover the highest primary containment vent capable of

rejecting all decay heat, and calculated the maximum primary containment

pressure capability.

In a report entitled " Calculation of Vent Flows for

the BSEP," dated July 29, 1988, the licensee reviewed four primary

containment vont flow paths and concluded that three of the four paths

would pass the anticipated design decay heat load.

Each of the three

acceptable paths vented the primary containment from the suppression

chamber. Although a vent path from an elevated location in the drywell

was not considered in the study, the licensee calculated the maximum

containment water level based on a vent path from the drywell (i.e.,

through valves V-9 ard V-10). The licensee indicated that the path was

equivalent to the suppression pool vent path and, therefore, was

technically adequate for not exceeding the maximum pressure limit;

however, a technical justification that the drywell vent path had

sufficient capacity to pass the decay heat load was not performed.

The calculated value for the maximum primary containment water level limit

was the elevation of the drywell vent elevation (i.e., 69.67 feet to the

center line of an 18-inch diameter vent pipe). A more conservative value

of 68.5 feet was used in the PSTG to ensure that water would not enter the

13

-

_ _ - _ _ _ _ _

.

'

.

'

.

vent piping and inhibit primary containment venting; however, this

conservatism was not included in the calculation.

The inspection team was concerned that the method of primary containment

water level measurement developed by the licensee did not have sufficient

accuracy to support controlling primary containment water level. The

Primary Containment Flooding Procedure, section 9, provided a method for

the operators to estimate the primary containment water level by using the

pressure instruments in the suppression chamber and at the bottom of the

drywell to trend the drywell pressure as a function of time during primary

containment flooding. Trending was required Secause the drywell pressure

instrument would be submerged at low primary containment water levels and

could not be used for measuring differential pressure and primary

containment watcr level. After adding the expected pressure head of the

water in the primary containment to the extrapolated pressure obtained

from trending, RPV injection was secured at the estimated total pressure

corresponding to the maximum primary containment water level. This

methodology was unreliable because it incorrectly assumed that the

pressure increase would be linear.

In addition, the inaccuracies involved

in this methodology would not support controlling primary containment

water level within an accuracy of 1.17 feet (i.e., the conservatism used

to prevent flooding the primary containment vent path).

The lack of primary containment water level instrumentation was noted

during the Detailed Control Room Design Review (DCRDR) in HED 206X-5093.

This deficiency will eventually be corrected by the installation of a

i

drywell pressure instrument above the maximum water level, thus supporting

i

accurate primary containment water level measurement.

Further licensee

I

action is necessary to revise the current procedures to ensure that the

1

primary containment water level measurement procedures can be implemented

l

l

effectively by the operators.

In addition, the new pressure instrumentation

j

should be installed as soon as possible.

I

2)

Worksheets WS-15 and WS-16 and plant-specific data package PSD-17

calculated the cold and hot shutdown boron weights required to poison the

reactor in the event of an ATWS.

In PSD-17, the licensee erroneously

calculatad the concentration of boron required due to several errors in

the conversion of the reference values provided by the vendor. This

incorrect conversion resulted in a calculation of the hot shutdown boren

weight which was 14.46 pounds too low. This incorrect value adversely

affected the calculations for: (1) time to inject boron (100 seconds

longer), (2) volume of the standby) liquid control (SLC) tank corresponding

to hot shutdown (68.5 gallons more , (3) SLC tank level indication for hot

shutdown (0.43 percent lower), and (4) the amount of borax required for

hot shutdown (127.6 pounds more). Although these errors resulted in

!

non-conservative values for the hot shutdown boron weights, the difference

!

(i.e., less that 5 percent) was unlikely to prevent the emergency shutdown

of the reactor due to the conservatism of the calculation.

Nevertheless,

~

these errors were not identified by the licensee's verification of the

i

calculation, including the independent verification by the NED. Further

'

i

licensee action is necessary to correct this error and ensure that all the

I

draft calculations are correct.

I

14

_

._

. _ _ _ _ _ _ - -

_ _ _ _ _ _____ - -

'

.

,

.

-

.

3)

Worksheet WS-AC4 detailed the calculation of the plant specific value for

i

drywell scram pressure. The numerical limit value was listed as 2.0 psig;

however, no calculation was provided to support the parameter. The basis

was listed as technical specifications 2.2.1-1 and 3.3-1 with an

amplifying comment that 2.0 psig was the scram setpoint for high drywell

pressure. As discussed in Section 3.3.1.1, the latter statement was

incorrect in that the high drywell pressure scram was set at 1.8 psig.

Further licensee action is necessary to ensure that the setpoint

documentation corresponds to values actually used.

4)

Worksheet WS-12 calculated the lowest suppression chamber pressure which

could occur when 95 percent of the non-condensables in the drywell had

been transferred to the suppression chamber.

A minor discrepancy was

identified in that the computed value was 13.07 psig, but the cover sheet

i

of the calculation indicated 13 psig without explanation.

PSTG, step

'

PC/P-1, also incorporated the value of 13.0 psig.

The PSTG should reflect

the calculations and any differences between the PSTG and the calculations

should be explained in the PSTG deviation documentation.

5)

Worksheet WS-8 calculated the highest suppression chamber pressure as a

function of the primary containment water level that would permit the

primary containment to maintain its pressure suppression function while

l

.

the RPV was at normal operating pressure.

Several administrative errors

I

l

that did not affect the technical adequacy of the calculation were noted.

'

Examples included differences between values which were transferred into

subsequent calculations.

l

3.2.4 Adequacy of Writer's Guide

A review of the PSWG was conducted to determine wbether it described acceptable

methods for accomplishing the objectives stated in NUREG-0899. The inspection

tet.m concluded that the PSWG was incomplete and should be supplemented with

detailed guidance in the following areas.

1)

Referencing Supporting Material - All figures, tables, and other

supporting materials that may be required in the performance of a

i

procedural step should be referenced explicitly in the E0P at the point at

!

which the information is needed.

For ex. ample, the "RPV Pressure Range for

System Operation Table," was not referenced or included in step 27 of the

End Path Procedure.

Similarily, although Primary Containment Control

Procedure, step PC/P-9, required controlling suppression chamber pressure

in the safe region of the pressure suppression pressure, no reference was

made in this step for the graph or figure to be used.

Guidance for

referencing supporting materials within the procedural steps should be

part of the PSWG.

2)

Referencing Other E0Ps - Several E0Ps directed the performance of a series

~Tsleps in accordance with other procedures.

In order to reduce

o

transition errors, the complete title of the procedure and its reference

,

number should be included in the procedural step.

In addition, a complete

technique that will aid the operator in making a correct identification of

these other procedures should be included in the PSWG.

l

!

I

15

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _

,

'

.

,

3)

Step Identification - The PSWG described a technique for identifying

critical action steps which required placing the symbol for a decision

block over the symbol of an action block.

This methodology was not an

effective method of identifying override requirements. As discussed in

Section 3.4.3.3, a critical step was overlooked during the simulator

exercise because the operator did not recognize that the critical step

represented an override condition. A more discernible shape coding

technique should be employed for deignating critical steps in the E0Ps.

4)

Operator Aids - Operating Instruction 01-41 discussed procedures and

methodologies for implementing operator aids at BSEP; however, this

instruction was not referenced by the PSWG.

Reference to this document,

including the basic criteria for design and control of operator aids,

should be incorporated into the PSWG.

The need for training operators on

the use of the operator aids should also be addressed.

5)

_ Color Coding - The PSWG contained no criteria for color coding the E0Ps;

however, the draft E0Ps employed a color coding scheme.

Guidelines and

direction on the uses of color should be included in the PSWG.

6)

Titles - The operators should be able to identify the basic scope of each

E0P by reading the title.

The E0P titles Path-1 through Path-5 failed to

indicate what the procedure was intended to accomplish.

Guidance for

constructing meaningful and unique titles for the E0Ps should be included

in the PSWG .

7)

Consistency of Step Numbering - Some steps within the E0P flow charts

applied the BWROG EPG convention for designating steps (i.e., PC/H-9),

while others employed a three digit system (i.e., 027).

A consistent

method for numbering the E0P steps should be incorporated and documented

in the PSWG.

3.2.5 Writer's Guide Implementation

The PSWG was independently verified to assess its implementation as a source

document for the preparation of the E0Ps. The verification process consisted

of comparing the E0P flowcharts and written procedures (e.g., LEPs, SEPs, etc.)

with the stated criteria and human factors guidance contained in the PSWG.

The

inspection team concluded that the PSWG was generally followed as a source

document for preparation of E0Ps; however, several minor deviations were

ider,ti fied.

Further licensee action is necessary to ensure that the criteria

and human factors guidance contained in the PSWG are reflected in the E0Ps.

1)

Instrument Accuracy - Some of the values referenced in the E0Ps could not

be obtained from the displays.

In Path-4 for example, the operators were

required to read the conductivity of the condensate booster pump to less

than 0.3 umhos. The instrument display, 1-00 CR-3075, did not support

this level of accuracy.

As demonstrated during the system walkthroughs,

the operators were unable to read the setpoint value of 0.3 mmhos from the

]

instrument scale. This deficiency was identified as a human engineering

deficiency (HED 20X5-5015) during the DCRDR; however, no corrective action

had been taken.

In addition, the resolution of the reactor building roof

j

radiation level instrument, CAC-AQH-1264-3, was unsuitable for reading the

i

E0P-specified setpoints of 3446 cpm (setpoint 1) and 4213 cpm (setpoint

16

- _ ___ _ _ _ _ _

'

.

'

,

'

.

2). Also, the E0P directed the operator to read the turbine building vent

radiation levels on instrument D12-RM-23; however, no setpoints were

identified on the instrument.

2)

Instrument Labels - The E0Ps referenced different units than those

inoicated on the instrument displays.

For example, the E0P referenced the

radiation level for service water effluent in units of counts per minute;

however, the instrument in the control room (i.e., D12-K805) for obtaining

this information was displayed in counts per second.

In addition, the

digital readout for monitoring stack releases, located on the control room

back panels, was not labeled and no units were identified.

Only the

!

I

value, 4.57E + 1, was displayed.

3)

location of Equipment - The E0Ps did not provide adequate location

information for specific equipment, controls, or displays.

For example,

the action steps to start the diesel fire pump, open all battery room exit

doors, or open emergency core cooling system (ECCS) pipe tunnel doors were

local operations.

The PSWG did not establish a standard method for

identifying the location of controls and displays external to the control

room.

4)

System Nomenclature - The E0Ps used inconsistent nomenclature for

equipment and systems.

For example, in the Level / Power Control Procedure,

steps 76 and 30, LPCI was used instead of RHR.

.

5)

Step Content - The E0Ps contained both decision and action steps or

contained more than one action or subject.

For example, in Path-3, steps

!

175 and 93, and Path-1, step 12, the decision step required several

actions on the part of the operator.

6)

Change Identification - There was no identification of the location of

recent changes in the written procedures. A change bar technique should

be used.

7)

Section Redundancy - PSWG, section 3.7, "Information/ Caution Steps," and

section 3.9, "Information Steps," appeared to be identical in content.

8)

Vocabulary - The E0Ps used verbs such as downrange, monitor, cycle, and

increase, which were not listed in Table 1 of the PSWG as approved verbs.

3.3 E0P Validation Using Plant Walkthroughs

In order to ensure that the E0Ps could be accomplished successfully, plant

walkthroughs for all the E0Ps and referenced operational procedures were

performed. The team verified that E0P instrument and control designations were

consistent with the installed equipment and that indicators, annunciators, and

controls referenced by the E0Ps were available to the operators. The

inspection team also verified the location and control of E0Ps-in the control

room. With the assistance of licensed operators, the team physically verified

that activities which would occur outside the control room during an accident

scenario could physically be accomplished and that tools, jumpers, and test

equipment were available to the operators.

17

- __

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

,

'

.

,

e

.

l

!

3.3.1 Technical Adequacy of Procedures __

The inspection team identified several deficiencies with respect to the

procedural completeness and technical adequacy of the E0Ps.

Although the

inspection team concluded that the operators could adequately perform the

procedures in spite of these deficiencies, further licensee action is necessary

'

to correct these deficiencies and perform an adequate verification and valida-

tion of the E0Ps.

1)

Path-1, steps 115 and 116, and Path-2, steps 165 and 166, directed the

operators to maximize the flow from the operating control rod drive (CRD)

l

system pumps by operating at the optimum pressure on the pump tead curve

,

The intent of these steps was to maximize CRD flow.

However, the steps

!

failed to accomplish the desired action because the operator was directed

I

to throttle the pressure control valve to maintain pressure equal to or

greater than 1000 psig and was never directed to increase CRD flow to the

reactor.

The steps should have directed the operator to maintain pressure

equal to or greater than 1000 psig but as low as possible.

Further

licensee action is necessary te modify these steps to ensure that the

intended action is accomplished.

l

2)

Primary Containment Flooding Procedure, section 7, step 3.d (2),

instructed the operators to lift wire number 25 from terminal 70, on

terminal board UU in control panel P601.

A note preceding the step stated

that the lead to be lifted was the lead entering from outside control

l

panel P601. The operator could not detennine which wire entered from

l

outside the panel because two wires with exactly the same number (i.e.,

l

number 25) were on the tenninal and both wires entered the same wireway.

l

This condition was noted at three other steps in the same section.

l

l

3)

Primary Containment Flooding Procedure, sections 7 and 8, step C.3, failed

'

to provide the operator with instructions concerning the level to which to

l

fill the primary containment.

4)

Primary Containment Flooding Procedure, section 2, did not list in the

note for manpower required the radwaste operator required to take several

actions necessary to support the evolution.

5)

Primary Containment Control Procedure, step PC/P-4, directed the operators

to vent the primary containment drywell through the standby gas treatment

(SBGT) system in accordance with Operational Procedure OP-10.

This

procedure only permitted venting the SBGT through two 1/2-inch lines

(valves V8 and V9) when drywell pressure was above 0.7 psig.

In this mode

of operation, the SBGT system vent path would have little or no effect on

controlling primary containment pressure.

The licensee should use the

10-inch ventilation damper (F-BFV-RP.) for venting, at least until the

pressure in the SBGT train reaches the limiting operating pressure.

6)

Primary Containment Control Procedure, steps PC/P-6 and PC/P-8, directed

the operator to initiate suppression pool and drywell sprays; however,

during the walkthroughs the operators were confused as to whether or not

to secure suppression pool sprays prior to initiating drywell sprays.

,

I

Augmented training or clarification in the E0P should be provided to

resolve this confusion.

18

_ ____ _____-___ _

-

f

,

)

.'

.

'

.

7)

Primary Containment Control Procedure, step SP/L-5.3, directed the

operator to drain the suppression pool to radwaste to control suppression

pool level.

This step did not provide alternate instructions if these

valves were interlocked closed from an isolation signal.

Further licensee

action is necessary to account for this possibility.

8)

Primary Containment Control Procedure, step SP/L-5.22 directed the

operators to maintain primary containment water level below 68.5 feet;

however, this step and subsequent steps did not reference the procedure to

accomplish this measurement.

9)

SEP-01, section 3, initiated primary containment venting before primary

containment pressure reached 70 psig by using preferentially listed vent

paths. After opening the proper valve, the subsequent action step

required continued venting of the primary containment if the initial

venting operation stabilized primary containment pressure below 70 psig.

The inspection team was concerned that the step provided inadequate

guidance to the operator concerning action required if the vent path was

more than adequate and primary containment pressure started to fall below

70 psig. The licensee should ensure that an approved PSWG action verb is

used which properly implements the intent of the BWROG EPG concerning

primary containment pressure control during venting.

10) SEP-04, steps 3 and 4, directed opening of reactor building inboard and

outboard ventilation isolation valves.

The terminology was incorrect in

that the procedure referred to the valves as reactor building inboard

(cutboard) isolation valve (s). The correct terminology was reactor

building vent inboard (outboard) isolation valve (s).

11) SEP-06, included entry conditions of drywell pressure which were below 2.0

psig.

The procedure was actually implemented when the shutdown cooling

interlocks were fulfilled at the corresponding drywell pressure of 1.8

'

psig. As discussed in Section 3.2.3.3, further licensee action is

necessary to ensure that procedural values are consistent with the plant

parameters used to initiate actions.

12) SEP-06 cautioned the operators concerning reactor power excursions when

the residual heat removal (RHR) system pumps were started in step C.48.

However, the correct reference for this precaution should have been a

subsequent action step which throttled open the injection valve. The

licensee should ensure that the caution correctly references the operator

action which actually affects reactor power level.

3.3.2 Availability of Special Tools and Equipment

The availability of special tools and equipment in the plant appeared to be

adequate to accomplish the activities required by the E0Ps. The team verified

that the plant equipment was accessible and available to perform the identified

task. A walkthrough was performed of the special tools and equipment used in

the E0Ps both in the control room and the plant.

Because the draft E0Ps had

not been implemented, not all equipment could be verified.

Nevertheless,

several specific examples were identified in which equipment or infonnation was

not available which could adversely affect the performance of the E0Ps and

their support procedures.

Based on the training and experience of the

19

'

.

'

.

operations staff, the inspection team concluded that the E0P actions could be

accomplished satisfactorily. However, based on the need to provide procedures

which can be implemented correctly by a newly qualified operator, and on the

guidance of NUREG-0899, the inspection team concluded that there was a

potential for operator confusion or error which could affect the performance of

the procedures.

Further licensee action is required to provide the necessary

equipment or information to ensure that operator confusion will not exist

during the performance of the procedures.

i

1)

LEP-02 provided an alternate control rod insertion method involving local

venting of the hydraulic control units (HCus). The venting operation used

control rod drive (CRD) vent hoses located in the toolbox on the 20-foot

elevation of the reactor building.

The toolbox contained two sets of

hoses with different types of connectors, only one of which would fit the

HCU vent block. The licensee could not determine the purpose of the

second set of hoses in the toolbox.

The inspection team was concerned

that in an emergency the presence of the incorrect hoses could delay the

performance of alternate control rod insertion. The inspection team also

noted that the toolbox did not contain any protective eouipment and that

the procedure did not warn the operators that HCU venting was a

potentially hazardous operation which could release contaminated, hot

reactor water.

In addition, the licensee indicated that the venting

procedure was a two-man job requiring one operator to perform the venting

operation in the overhead while a second operator coordinated the

activities with the control room and verified that the correct hydraulic

i

control unit was being vented from below.

However, the procedure only

required the resources of one operator to perform the venting operation.

2)

The Primary Containment Flooding Procedure required the use of several

electrical jumpers. Generic jumpers were available to the operators to

perform the E0P actions; however, these jumpers had closed-end

terminations. The use of closed-ended jumpers required the operator to

(

remove the terrrinal screw, install the additional terminal, recapture all

i

terminals, and re-install the terminal screw. The inspection team was

concerned that this task was unnecessarily complex for emergency

conditions.

The use of open-ended terminations, which could be slipped

under a loosened screw, would simplify the task.

In addition, the

inspection team noted that the procedure lacked direction concerning

insulation of lifted leads, and that insulating materials were not readily

,

available.

The inspection team also observed that some electrical relays had wiring

diagrams posted adjacent to the relays to aid the operators in identifying

the terminal locations; however, not all relays used in the E0Ps were

identified in this manner.

Further licensee action is necessary to

provide installation specific jumpers for use in accomplishing the E0P

action steps and to provide consistent use of operator aids for the

identification of relay terminal locations.

,

3)

During the E0P simulations, the control room operators directed the

auxiliary operators to perform numerous actions in the plant.

For

example, steps 67 and 68 in Path-1, required opening battery room exit

doors and ECCS pipe tunnel doors. These actions were initiated by the

control room operator using the public address (PA) system and required

20

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

'

,

.

'

.

the auxiliary operators to find a PA station to report the completion of

the directed actions.

Alternate communications techniques, such as

hand-held radios, were not available for communicating with the operators

perfonning local actions. The inspection team concluded that the licensee

should give further consideration to the use of hand-held radios to permit

reliable communication with the control room under emergency conditions.

i

3.3.3 Station Material Condition

The inspection team reviewed the material condition of the station during the

plant walkthroughs and ensured that necessary equipment and components were

dCCessible and functional.

The overall material condition of the plant

appeared good. The team did not observe any interferences in the reactor

building which would adversely affect emergency actions. The inspection team

noted that significant amounts of non-combustible material were located in the

bottom of control panel P601; however, the licensee initiated corrective action

to clean the panel and inspected and cleaned other panels as required.

The

team verified that emergency lighting was available for E0P operator actions

and noted that lighting was available within electrical cabinets requiring

terminal manipulations.

However, during the walkthroughs, the operators would

not operate the switches to turn on the lights in the cabinets because the

switches were not labeled.

Further licensee action is necessary to correct

this deficiency.

3.3.4 Reactor Building Accessibility

The licensee performed a design review entitled, " Post-Accident Control of

Radiation in Systems Outside Containment of PWRs and BWRs," to meet the

requirements of paragraph 2.1.6 of NUREG-0578, "TMI-2 Lessons Learned Task

Force Status Report and Short Term Recommendations." The inspection team noted

that the extent of the design review fulfilled the additional requirements of

NUREG-0737, paragraph II.B.2, concerning the same subject.

The inspection team

evaluated the results of this design review and its impact on the ability of

operators to perform the emergency actions of the E0Ps.

The ability of the operators to perform the E0P actions successfully would be

dependent on access to the reactor and radwaste buildings. Access to the

reactor building was dependent on the specific accident scenario, and access to

the radwaste building was dependent on the location of primary system leakage.

Although the licensee's radiation protection procedures allowed operator entry

into high radiation level areas under the supervision of radiation protection

personnel, the E0P contingency actions could not be performed if radiation

levels prevented entry.

The design review was based upon the source terms

specified by Pegulatory Guides 1.3 and 1.4 and the accidents of Chapter 14 of

'

the BSEP Final Safety Analysis Report (FSAR). The design review concluded that

entries into unprotected areas or areas with prohibitively high dose rates

would not be required for mitigation of the accidents. However, several areas

were identified which could require operator entry during recuvery operations.

The inspection team concluded that multiple methods of implementing the E0P

contingency actions had been adequately considered in the development of the

E0Ps. However, the inspection team identified two actions, during the

walkthrough of the plant, for which an alternative method of accomplishment had

l

21

_-_

- _ _ _ _ _ _ _ - _ _ -

_- -

.

t

.

-

.

not been adequately considered.

Further licensee action is necessary to

correct these discrepancies.

1)

The LEP-01 and Primary Containment Flooding Procedure identified several

local operator actions to inject service water and demineralized water

into the RPV.

These actions included opening the residual heat removal

(RHR) loop cross tie valve, Ell F010, in the high pressure core injection

(HPCI)systemmezzaninearea.

This valve was a normally de-energized

motor-operated valve whose breaker was removed from its cubical to ensure

separation of the two trains of RHR.

A significant amount of time was

required to operate this valve manually in area in which radiation levels

could be as high as 20000 R/HR one hour into an accident.

Since this

valve had the potential to be operated remotely, further licensee

consideration should be given to reinstalling the valve breaker rather

than attempting manual operation.

2)

SEP-06, step C.24, required the operator to monitor the RHR heat exchanger

outlet conductivity at a local instrument in the south RHR room.

This

,

area would have extremely high radiation levels in the accident conditions

during which performance of the step would be required.

The inspection

team noted that control room panel alarm, A-03, tile 2-10, monitored the

desired location and alarmed at the value specified in the E0P (i.e., 10

umho/cm).

Further licensee action is necessary to ensure that remote

instrumentation is used where possible in lieu of local monitoring in high

radiation areas.

i

3.4 E0P Validation Using Plant Simulator

j

To ensure that the E0Ps could be implemented correctly under emergency

j

conditions, the inspection team developed and performed four accident scenarios

utilizing licensed operators. The accident scenarios determined whether the

)

E0Ps provided the operators with sufficient guidance and clearly outlined their

required actions during an emergency; verified whether the E0Ps caused the

operators to interfere physically with each other; verified that the procedures

did not duplicate operator actions unless required; and verified that

transitions from one E0P to another or to other procedures were accomplished

satisfactorily.

3.4.1

Scenario Description

The first scenario involved a rupture of the feedwater pump suction header from

100% power with a spurious group 1 isolation signal inserted at the time of the

reactor scram due to low RPV water level.

The SRVs opened on high RPV pressure

following the main steam isolation valve (MSIV) closure.

One safety relief

valve (SRV) stuck open and remained open throughout the scenario. One minute

after the scram and MSIV isolation, a small steam leak was initiated into the

drywell . The high pressure core injection (HPCI) system, pump B of the control

rod drive (CRD) system, and the loop B heat exchanger of the residual heat

removal (RHR) system were out of service throughout the event.

Following the

reactor scram, the operators performed Path-4 when RPV water level decreased

below +112 inches.

The operators exited Path-4 and performed steps RC/L and

RC/P of the End Path Procedure concurrently to restore RPV water level and

pressure. The operators performed the Primary Containment Control

Procedure to control suppression pool temperature and drywell pressure and

22

_

___

_______________ _-

--

__

.-

.

4

i

temperature, and depressurized the RPV in accordance with the End Path

i

Procedure, when drywell temperature exceeded 300 degrees F.

The second scenario exercised the Level / Power Control Procedure with alternate

l

boron injection. A spurious group 1 isolation signal initiated the event and

resulted in a failure of all control rods to scram.

Failure of the standby

liquid control (SBLC) system to inject along with both reactor water cleanup

(RWCU) system pumps being out of service required the use of the Alternate

Boron Injection Procedure using the CRD system. A small break loss of coolant

accident (LOCA) in the drywell required emergency depressurization when drywell

temperature exceeded 300 degrees F.

The scram condition required the

performance of Path-1 and eventually the Level / Power Control Procedure for the

l

ATWS condition. The Primary Containment Control Procedure was used to control

drywell and suppression pool temperature and pressure.

The third scenario exercised the Secondary Contairiment Control Procedure and

i

the Radioactive Release Control Procedure. A loss of feedwater resulted in a

reactor scram on low RPV water level coincident with a fuel element failure.

Maintenance. activities in the HPCI room required the reactor core isolation

cooling (RCIC) to HPCI room door to remain open to allow passage of hoses.

When the RCIC system started on low RPV water level a steam leak occurred at

the RCIC steam inlet valve (F0-45). The steam leak caused a RCIC system

isolation signal.

The RCIC steam supply containment isolation valves failed to

isolate and caused a HPCI isolation signal several minutes later due to the

open door between the two rooms. The scram coincident with an RPV level below

l

+112 inches required performance of Path-4. The radioactive steam leak in

'

secondary containment required performance of the Secondary Containment Control

Procedure.

Exceeding the reactor building roof vent annunciator setpoint

required performing the Radiological Release Control Procedure. When the

operators determined that more than one area had exceeded its maximum safe

operating radiation level, the End Path Procedure required emergency

depressurization.

l

The fourth scenario required venting primary containment to control primary

.

containment hydrogen concentrations.

RHR loop B was out of service throughout

l

emergency bus (E-3) ge break LOCA coincident with a failure of the 4160 vcit

the scenario. A lar

initiated a reactor scram and an ECCS actuation. The A

loop RHR injection valve failed to open, leaving only one core spray (CS) pump

available for injection.

The reactor core was uncovered, resulting in fuel

,

damage and the release of hydrogen to the primary containment. The scram with

high drywell pressure required performance of Path-5 and the Primary

Containment Control Procedure.

The loss of power to emergency bus E-3

unexpectedly resulted in the inability of the operators to perform the primary

containment venting procedure because the torus purge exhaust valve, CAC V-8,

was powered from emergency bus E-3.

Further licensee action is necessary to

ensure that an alternate method is available to vent the primary containment

ouring a partial loss of power condition.

3.4.2 Limitations of the Plant-Specific Simulator

The plant-specific simulator located on-site was used for the E0P scenarios.

The simulator demonstrated extremely poor modeling with respect to decay heat

and RPV water level response.

For example, during scenarios in which all high

pressure injection had failed and with mass being removed by open SRVs or a

23

1

f

'

,, .

.

.

small break LOCA, RPV water level would continue to increase.

Following

MSIV isolations from 100 percent power with end of life (COL) decay heat

loading and no steam being drawn off by the HPCI or RCIC systems, it was not

necessary to use the SRVs to control RPV pressure.

In fact, RPV pressure would

decrease with no external energy removal in progress. As a result, the RC/L

steps of Path-1 were not able to be simulated past the initial entry steps.

The inspection team concluded that the plant-specific simulator was not.an

effective tool for operator training on the Level / Power Control Procedure, E0P

Path-5, or any of the E0P steps requiring level control manipulations. As

previously discussed in Section 3.1.2, the simulator modeling deficiencies also

adversely affected the ability of the licensee to perform validation for any

E0P steps which required level manipulations.

3.4.3 Observations and Conclusions

The inspection team concluded that the operating crew could satisfactorily-

implement the E0Ps to shutdown the reactor and return the plant to a safe,

stable condition. Overall, the operators performed well and demonstrated a

good understanding of the E0Ps which was indicative of a high level of training

on the procedures. As discussed in Section 3.1.1, the inspection team

concluded that the timely implementation and execution of the E0Ps required the

active participation and assistance of the STA because the licensee included

the post-trip actions in the E0Ps and developed overly complex procedures. The

inspection team identified concerns in the following two areas.

1)

Control Room Responsibilities - During all four scenarios, the inspection

team observed that the shift foreman (SF) directly supervised the two

,

control operators and directed the performance of the E0Ps and that the

l

shift operating supervisor (SOS) monitored the emergency plan and

performed the required notifications.

The shift technical assistant (STA)

monitored the emergency response facility information system (ERFIS) and

available control room indications for key parameters and trends.

In

I

addition, the STA monitored changing plant conditions to identify E0P

J

entry conditions and to advise the SF regarding the required actions. The

inspection team also noted that the STA performed E0P steps in legs which

the SF did not have time to execute.

This was particularly evident in the

<

third scenario involving the Secondary Containment Control Procedure.

1

The BSEP administrative instructions required the STA to provide an

overview of the plant conditions and ensure that all the required E0P

steps were completed.

In actual practice, the STA independently performed

portions of the E0Ps in order to provide more time for the SF to read and

complete the post-trip scram recovery actions of the E0Ps.

The inspection

team concluded that the level of detail of the BSEP E0Ps did not permit a

single individual sufficient time to direct the performance of all

required actions of the E0Ps.

The inspection team also observed that the SF was not able to perfom all

the parallel steps as required by the BWROG EPGs. This was clearly

demonstrated in the first scenario involving the performance of the

Primary Containment Control Procedure.

During the scenario, the SF

completed only two steps of the five required parallel flowpaths (i.e.,

DW/T and PC/P).

The remaining three flowpaths (i.e., SP/T, SP/L, and

PC/H) were not performed. Another example occurred in the third scenario

24

- _ _ _ - - _ _ _ _ _ _ _ - - _ - _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ -

_ _ _ _ _ ______ _ __

- _

_ _ - _ _ - _ _ _ - _ _ _ - _ - _ _

,,

.

+

.

involving the Secondary Containment Control Procedure.

During this

scenario, the SF completed only one of the three required parallel

flowpaths (i .e. , SC/R).

In this example, the SF directed the control

operator to obtain the area radiation levels from the back panel, but not

the area temperature and level readings. The failure to execute all legs

-

of the E0Ps potentially prevents monitoring and control of all symptoms

indicative of an accident condition. As discussed in Section 3.1.1,

further licensee action is necessary to (1) accurately define and

implement the control room responsibilities of the STA and SF during E0P.

performance, (2) remove the event-based actions from the E0Ps, and (3)

reduce the level of complexity of the E0Ps.

2)

Critical Action Steps - Critical action step RR-5 in the Radioactive

-

,

Release Control Procedure required the operators to execute the subsequent

)

actions of the flowpath only if an ALERT was not declared as a result of a

i

radioactive release. As discussed in Section 3.1.1, these actions were

l

l

event-based and not appropriate for inclusion in the E0Ps. Durir.g the

l

third scenario, the_ STA and SF performed these action steps after an ALERT

had been declared. Although in direct conflict with the procedural

requirement of step RR-5, the licensee's training staff indicated that it

was desirable to perform these action steps even after an ALERT had been

declared.

Further licensee action is required to correctly train and

implement the critical actlon of step RR-5.

During the third scenario, the operators incorrectly performed emergency

,

depressurization in accordance with the End Path Procedure because they

!

missed critical action step 64 of the Level / Power Control Procedure. As a

result, the operators bypassed the cautions concerning power oscillations

during an ATWS contained in the procedure.

Although, the SF correctly

i

controlled injection flows and reactor power level, he subsequently

indicated that he did so as a result of his previous training and had

missed the precautions of the critical action step.

'

The operators used a marker to maintain peacekeeping within the E0Ps and

to note critical action steps and, as a result, were able to explain

accurately where they were in each of the flowcharts.

However, the

scenarios demonstrated that they experienced difficulty in identifying and

'

,

monitoring the override requirements of critical action steps.

Because

missing a critical action step has a significant potential to result in

.

severe core damage, the inspection team concluded that further licensee

l

action is necessary to identify, train, and procedurally support a more

effective method for monitoring the critical action steps.

3.5 Operator Interviews

i

The inspection team conducted interviews with three shift operating

-'

supervisors, four control operators, and one auxiliary operator. These

interviews developed information on the effectiveness of the E0Ps and did not

examine the qualifications of the operators.

Each interview lasted

!

approximately one hour. The following observations summarize the comments

i

volunteered by the operators.

25

__- _ _ - _ - _ _ _- __-_ _ ____-__ _ ___

___

..

.

e

.

.

3.5.1 Observations and Conclusions

1)

Equi ament Design - The operators experienced difficulty in locating and

reacaing several valves outside the control room.

For example, the

operators suggested cutting a manhole in the grate that covered the

condensate header valve, C0-V304, to enhance the accessibility from above.

In addition, the operators also suggested implementing hardware

modifications to make the valve more easily accessible and labeling the

.

!

RCIC CST suction valve, CO-V301, on a nearby wall to clarify its location.

Although the E0P provided location information, the operators indicated

)

that the use of signs would' aid performance of the E0Ps.

2)

Assignment of Duties - The E0Ps clearly defined the number and

qualifications of operations personnel required for executing the E0Ps.

Major tasks and duty assignments were clearly delineated and unambiguous.

j

The operating instructions delineated the basic philosophy and established

i

practices for personnel assignments.

l

3)

0)erator Training - All SFs and C0s had received preliminary training on

tie use of the draft EOPs. Approximately two weeks of combined classroom

l

and simulator training were devoted to the use of E0PS; however, formal

!

training on the E0Ps for the A0s had not yet been accomplished.

The

operators indicated that additional training was scheduled before the

draft E0Ps would be implemented.

Ir general, the operators considered

their training on the E0Ps to be adequate; however, more training would be

beneficial. Some operators expresseii concern regarding the transfer of

training betweer. the new procedures end the old procedures.

4)

Validation and Verification of E0Ps - The verification and validation of

the E0Ps included a combination of system walkthroughs and simulator

exercises.

In addition, operator training accomplished portions of the

l

verification and validation process.

For example, verification of the

technical adequacy for selected E0Ps was performed during classroom

'

discussions.

.

5)

System for Making Changes to E0Ps - A fonnal system existed for making

changes to the E0Ps.

The operators submitted changes to the E0Ps in

accordance with Operating Instruction 01-28.

6)

Calculations - The E0Ps required the operators to perform very few

calculations and did not require complex calculations.

7)

E0P Availability - All the E0Ps were located within the control room and

i

were inraediately accessible by the operators. All of the operators

'

reported that there were no problems in locating and retrieving the

required E0Ps needed to perform a spccific function.

Nevertheless, the

inspection team believed that further consideration should be given to

locating the E0Ps which would be required to be performed outside the

control rocm at a locally accessible area.

NVP.EG-0899 required that the

procedures be available at all locations in the plant where equipment is

to be manually operated under emergency conditions.

8)

Communications - The operators considered the communications inside the

control room to be adequate and reported no conditions where it was hard

,

26

l

_

.i

,

.

'

.

to hear or convey verbal instructions in the control room. All operators

expressed the need to keep the number of personnel in the control room to

a minimum during en emergency. The operators identified that communica-

tions would be difficult in the diesel building and the RHR pump room (-17

level) during an emergency. The inspection team noted that communications

from outside the control room were only available through the PA system

and that the availability of radios as an alternative mode for communica-

tions would be a valuable asset.

3.6 Primary Containment Venting Provisions

The inspection team reviewed the " Primary Containment Venting Procedure,"

E0P-01-SEP-01, to determine the adequacy of the procedure and the feasibility

of the vent paths.

The inspection team also reviewed the results of the

special Probabilistic Risk Assessment based operational safety inspection

cor. ducted by the NRC in March 1988. The inspection team performed a

l

walkthrough of all primary containment vent paths which had not previously been

l

examined during the earlier inspection, and verified that all necessary

equipment was available.

The Primary Containment Control Procedure initiated venting of the primary

containment, irrespective of the off-site release rate, for conditions of high

l

pressure (i.e., 70 psig in step PC/P-12) and for conditions of high hydrogen or

i

oxygen (i.e., 6 percent and 5 percent, respectively, in step PC/H-16). The

i

shift foreman had the final authority for venting the primary containment under

these conditions.

4

The licensee had established hard pipe vent paths which were capable of

l

removing the decay heat load required by Revision 4AF of the BWROG EPGs.

I

E0P-01-SEP-01 preferentially listed the vent paths from the small bore pipe to

the large bore piping, to control the primary containment pressure. All the

vent paths were monitored release paths that permitted off-site dose

calculations to be performed. Although the vent paths used hard piping, low

pressure ducting was installed at transitions to the standby gas treatment

(SBGT) system and the reactor building purge exhaust system fans. A recent

study completed by the licensee concluded that the pressure at the fan duct

work could exceed acceptable limits and a further evaluation was in progress at

the time of inspection. This evaluation should be completed in a timely manner

by the licensee.

The inspection team was also concerned about the ability of the operators to

establish a vent path during reduced power capability or station blackout

conditions. As discussed in Section 3.4.1, the inspection team noted during

the simulator exercises that the operators were unable to establish a vent path

to remove simulated excessive hydrogen with the loss of one division of

essential power.

Contingency plans were under development by the licensee for

the conditions of loss of power, including containment venting provisions.

This effort should be completed expeditiously by the licensee.

4.0 MANAGEMENT EXIT MEETING

The inspection team conducted an exit meeting on October 7, 1988, with licensee

management.

During this meeting, the inspection team identified the inspection

findings and provided the licensee with an opportunity to question the

observations. The inspection team also detailed the scope of the inspection

27

I

f.,*

,

'

,,

'

e

and informed the licensee of the conclusions identified in this report. Mr.

l

Jim Konklin, Section Chief, Special Team Support and Integration Section,

Office of Nuclear Reactor Regulation, and Mr. Caudie Julian, Branch Chief,

Operations Branch, Region II, represented NRC management at the final exit

meeting. Appendix A identifies the licensee personnel who participated in this

meeting.

l

l

1

l

l

28

_ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ -

'

.'

,

'

'

.,

'

e

APPENDIX A

PERSONNEL CONTACTED

A large number of people, including the following licensee personnel, were

contacted during the inspection.

I

• J. Harness, Plant General Manager

• K. Enzor, Director, Regulatory Compliance

,

• L. Jones, Director, Quality Assurance and Quality Control

i

• C. Blackmon, Manager, Operations

A. Hegler, Superintendent, Operations

• W. Martin, Principal Engineer, On-site Nuclear Safety

• J. Titrington, Principal Engineer, Operations

• M. Sawtschenko, Operations

S. Reynolds, Operations

J

M. Amato, Operations

l

• M. Williams, Senior Specialist, Operations

l

D. LaBelle, Shift Supervisor, Operations

l

M. Schall, Shift Foreman, Operations

l

E. Hutt, Shift Foreman, Operations

l

l

K. Chism, Shift Foreman, Operations

i

K. Horn, Shift Foreman, Operations

'

R. Gibbs, Shift Technical Advisor, Operations

l

H. Harrelson, Operations

R. Mullis, Operations

D. Best, Operations

B. Jones, Operations

D. Jenkins, Operations

R. Blair, Operations

R. Knight, Operations

R. Poulk, Regulatory Compliance

• T. Jones, Regulatory Compliance

• J. Moyer, Manager, Training

l

E. Hawkins, Training

• M. Shealy, Project Specialist', Training

• B. Strickland, Project Specialist, Operations

• A. Schmich, Senior Specialist, Corporate Nuclear Licensing

• Denotes those personnel present at the exit meeting on October 7, 1988.

I

A-1

,

h

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

&*.

.

I

a

,,

'

e

APPENDIX B

l

DOCUMENTS REVIEWED

l

l

Emergency Procedure Guidelines (EPGs), Revision 4AF, March 1987

l

Plant Specific Technical Guideline (PSTG) for EPG Revision 4, Draft D

i

EPG/PSTG Step Documentation, Draft D

l

Appendix A PSTG/EOP Step Documentation, Draft C

l

Procedures Generation Package (PGP), August 17, 1983

Administrative Instruction AI-95, " Verification and Validation Program for

EPG, Revision 4, based Emergency Operating Procedures," Draft A

MST-RPS-26R, "Drywell Pressure Setpoint Calibration," Revision 2

Engineering Evaluation Report No. 85-0231, Revision 0

l

General Area Personnel Dose Rates Versus Time (post-LOCA)

l

Emergency Operating Procedures (E0Ps):

E0P-01-UG, " User's Guide," Draft B

l

E0P-01-FP-1, " Path-1," Draft 0

l

E0P-01-FP-2, " Path-2," Draft E

E0P-01-FP-3, " Path-3," Draft E

E0P-01-FP-4, " Path-4," Draft D

l

E0P-01-FP-5, " Path-3," Draft D

E0P-0 rPP-5, "End Path Procedure," Draft H

E0P-01-LPC-1, " Level / Power Control Procedure," Draft E

I

E0P-02-PCCF, " Primary Containment Contml Procedure," Draft F

'0P-03-SCCF, " Secondary Containment C

.rol Procedure," Draft G

l

,

l

'

E0P-04-RRCP, " Radioactivity Release Co.. trol," Draft D

E0P-01-ALC, " Alternate Level Control," Revision E

i

E0P-01-AEDP, " Alternate Emergency Depressurization Procedure,"

Revision D

E0P-01-StCP, " Steam Cooling Procedure," Revision A

E0P-01-FP, "RPV Flooding Procedure," Revision E

E0P-01-PCFP, " Primary Containment Flooding Procedure," Revision B

E0P-01-LEP-01, " Alternate Coolant Injection," Revision 005

E0P-01-LEP-02, " Alternate Control Rod Insertion," Revision 005

E0P-01-LEP-03, " Alternate Boron Injection," Revision 004

E0P-01-SEP-01, " Primary Containment Venting," Draft D

E0P-01-SEP-02, "Drywell Spray Procedure," Draft C

.

E0P-01-SEP-03, " Suppression Pool Spray Procedure," Draft C

'

E0P-01-SEP-04, " Reactor Building HVAC Restart Procedure," Draft C

E0P-01-SEP-05, " Primary Containment Purging," Draft C

E0P-01-SEP-06, " Shutdown Following Boron Injection," Draft B

E0P-01-SEP-07, " Bypassing RWCU Filter Domineralizers," Draft B

E0P-01-SEP-09, "CRD Flow Maximization," Draft B

Operating Instructions and Procedures:

01-28, " Appendix C Writer's Guide for Emergency Operating Procedures

(EOPs)," Revision 6

01-37, " Preparation and Review of the Plant-Specific Technical Guideline

i

for EPG Revision 2,"

Revision 001

B-1

'

_ _ _ _ _ _ _ _ _ - _ _ _ _-_ -

l

1

,

g,*

,,

'

a

PT-16.2, " Primary Containment Volumetric Average Temperature,"

Revision 20

CP-05, " Unit Shutdown," Revision 28

OP-10. " Standby Gas Treatnient System," Revision 32 (Unit 2)

OP-17, " Residual Heat Removal System," Revision 77 (Unit 2)

OP-24, " Containment Atmosphere Control," Revision 26 (Unit 1)

i

,

)

B-2

L-______-____-_________________-

_ _ _ _ _ _